Simulation of infrared, radar, and other imaging sensors plays an important role in the planning and rehearsal of military missions and in the training of mission personnel. The challenge is to develop technology that...
详细信息
ISBN:
(纸本)0819412015
Simulation of infrared, radar, and other imaging sensors plays an important role in the planning and rehearsal of military missions and in the training of mission personnel. The challenge is to develop technology that will support the rapid use of reconnaissance imagery to generate cockpit sensor displays that accurately represent the real world while insuring correlation among the out-the-window scenes and sensor displays. This paper describes a novel, neural-network-based technique for infrared and radar image simulation directly from multi-spectral imagery (MSI). Source imagery, its processing using neuralnetworks, and infrared and radar image simulation results are presented. Issues related to MSI database generation are also described.
Adaptive spline interpolation (which is equivalent to the use of a type of radial basis function neural network) is investigated for digital image interpolation (i.e., for resolution enhancement). Test image results i...
详细信息
ISBN:
(纸本)0819439835
Adaptive spline interpolation (which is equivalent to the use of a type of radial basis function neural network) is investigated for digital image interpolation (i.e., for resolution enhancement). Test image results indicate that adaptive spline interpolation of a low-resolution image is superior to non-adaptive interpolation if the adjustable parameters are chosen to yield the best match to a known object in a corresponding high-resolution image.
This two-volume-set (CCIS 188 and CCIS 189) constitutes the refereed proceedings of the International conference on Digital Information processing and Communications, ICDIPC 2011, held in Ostrava, Czech Republic, in J...
ISBN:
(数字)9783642224102
ISBN:
(纸本)9783642224096
This two-volume-set (CCIS 188 and CCIS 189) constitutes the refereed proceedings of the International conference on Digital Information processing and Communications, ICDIPC 2011, held in Ostrava, Czech Republic, in July 2011. The 91 revised full papers of both volumes presented together with 4 invited talks were carefully reviewed and selected from 235 submissions. The papers are organized in topical sections on network security; Web applications; data mining; neuralnetworks; distributed and parallel processing; biometrics technologies; e-learning; information ethics; imageprocessing; information and data management; software engineering; data compression; networks; computer security; hardware and systems; multimedia; ad hoc network; artificial intelligence; signal processing; cloud computing; forensics; security; software and systems; mobile networking; and some miscellaneous topics in digital information and communications.
Cellular neuralnetworks (CNN) provides fast parallel computational capability for imageprocessingapplications. The behavior of the CNN is defined by two template matrices. In this paper, adjustment of these templat...
详细信息
ISBN:
(纸本)0819453625
Cellular neuralnetworks (CNN) provides fast parallel computational capability for imageprocessingapplications. The behavior of the CNN is defined by two template matrices. In this paper, adjustment of these template-matrix coefficients have been realized using supervised learning algorithm based on back-propagation technique and wavelet function. Back-propagation algorithm has been modified for dynamic behavior of CNN. Wavelet function is utilized to provide the activation function derivation in this learning algorithm. The supervised learning algorithm is then executed to obtain a compact CNN architecture, called as Wave-CNN. The proposed new learning algorithm and Wave-CNN architecture performance have been tested for 2D imageprocessingapplications.
Accelerating deep neuralnetworks (DNNs) has been attracting increasing attention as it can benefit a wide range of applications, e.g., enabling mobile systems with limited computing resources to own powerful visual r...
详细信息
ISBN:
(纸本)9781577358008
Accelerating deep neuralnetworks (DNNs) has been attracting increasing attention as it can benefit a wide range of applications, e.g., enabling mobile systems with limited computing resources to own powerful visual recognition ability. A practical strategy to this goal usually relies on a two-stage process: operating on the trained DNNs (e.g., approximating the convolutional filters with tensor decomposition) and fine-tuning the amended network, leading to difficulty in balancing the trade-off between acceleration and maintaining recognition performance. In this work, aiming at a general and comprehensive way for neural network acceleration, we develop a Wavelet-like Auto-Encoder (WAE) that decomposes the original input image into two low-resolution channels (sub-images) and incorporate the WAE into the classification neuralnetworks for joint training. The two decomposed channels, in particular, are encoded to carry the low-frequency information (e.g., image profiles) and high-frequency (e.g., image details or noises), respectively, and enable reconstructing the original input image through the decoding process. Then, we feed the low-frequency channel into a standard classification network such as VGG or ResNet and employ a very lightweight network to fuse with the high-frequency channel to obtain the classification result. Compared to existing DNN acceleration solutions, our framework has the following advantages: i) it is tolerant to any existing convolutional neuralnetworks for classification without amending their structures;ii) the WAE provides an interpretable way to preserve the main components of the input image for classification.
In this paper, we present the performance analysis of three different neural network paradigms, ART-1, ARTMAP inspired ART-1 and Neocognitron, for part recognition in an autonomous assembly system. This intelligent ma...
详细信息
ISBN:
(纸本)0819412015
In this paper, we present the performance analysis of three different neural network paradigms, ART-1, ARTMAP inspired ART-1 and Neocognitron, for part recognition in an autonomous assembly system. This intelligent manufacturing system integrates machine vision, neuralnetworks and robotics in order to identify, locate and assemble randomly places components on printed circuit boards requiring precision assembly. The system uses an IBM 7547 robot controlled by an IBM PS/2 computer, a CCD camera and an image capture card. The electronic components are identified and located by using artificialneuralnetworks. The system's component location and identification accuracy are tested on all test components. The results show that the neocognitron-based system performed better than the other two systems.
In this article, a system for automatic recognition of selected road users using artificialneuralnetworks is proposed. Five models of neuralnetworks were tested with various configurations. With the special prepare...
详细信息
ISBN:
(数字)9788362065424
ISBN:
(纸本)9788362065424
In this article, a system for automatic recognition of selected road users using artificialneuralnetworks is proposed. Five models of neuralnetworks were tested with various configurations. With the special prepared image database, training, validation and tests were conducted to identify five classes of road users: cyclists, people on electric scooters, roller skates, pedestrians and people using personal transport devices. A web application that recognize these classes of road users was also prepared.
Complex valued artificialneuralnetworks (CVANN);is a kind of artificialneural network of which parameters such as weight, threshold, input and output consisting of complex numbers. In recent years, CVANN has increa...
详细信息
ISBN:
(纸本)9781467373869
Complex valued artificialneuralnetworks (CVANN);is a kind of artificialneural network of which parameters such as weight, threshold, input and output consisting of complex numbers. In recent years, CVANN has increasing interest because CVANN gives better results in nonlinear signal processing and imageprocessing problems. There has been no software introduced in the literature for CVANN yet. In this study, CVANN software (CV-ANN), which is developed with object-oriented programming language C#, is presented. In this software tool which is designed with a user-friendly interface, user can set the parameter values of CVANN. Changes in neural network output with respect to entered parameter values are showed graphically. Developed software can facilitate the work of researchers in academic and scientific studies. At the same time, this software can be useful as an educational tool in the sense of making the working mechanism of CVANN clearer. CV-ANN is tested on XOR problem which is a standard test and CV-ANN proves its reliability performing the classification with the accuracy obtained by the same algorithm in the literature.
In this study, a novel unsupervised incremental neural network is proposed for the segmentation of remote sensing images. Feature vectors are formed by the intensity of one pixel of each channel. The trainning set of ...
详细信息
ISBN:
(纸本)0780383184
In this study, a novel unsupervised incremental neural network is proposed for the segmentation of remote sensing images. Feature vectors are formed by the intensity of one pixel of each channel. The trainning set of DAYS network is formed by using all pixels of the image. The remote sensing image is segmented according to the decision of the network. In the study, the segmentation results of DAYS and Kohonen networks are compared
Fuzzy neuralnetworks is gaining much researcher's interest and have attracted considerably attention recently, due to its diverse applications in such fields as pattern recognition, imageprocessing and control. ...
详细信息
ISBN:
(纸本)0780376900
Fuzzy neuralnetworks is gaining much researcher's interest and have attracted considerably attention recently, due to its diverse applications in such fields as pattern recognition, imageprocessing and control. However, this type of neural system, as same with that of multilayer perceptron, has a drawback due to its huge neural connections. In this article we proposed a method for optimizing the structure of a fuzzy artificialneuralnetworks (FANN) through genetic algorithms. This genetic algorithm (GA) is used to optimize the number of weight connections in a neural network structure, by evolutionary calculating the fitness function of those structures as individuals in a population. The developed optimized fuzzy neural is then applied as the pattern recognition in our odor recognition system. Experimental results show that the optimized neural system provides higher recognition capability compare with that of unoptimized neural system. Recognition rate of the unoptimized neural structure is 70.4% and could be increased up to 85.2% in the optimized neural system. It is also shown that the computational cost of the optimized structure of neural system is also lower than the unoptimized structure.
暂无评论